National Computational Infrastructure

NCI

Specialised queues

In addition to the standard Raijin compute nodes, you can also access each of the following specialised queues.

hugemem:

  • 2 x 14 cores (Xeon CPU E5-2690 v4, 2.6 GHz) in 10 compute nodes
  • 1 TBytes of RAM
  • 420 GB of local jobfs
  • charge rate, 1.25SU per cpuhour
  • minimum number of ncpus request is 7, must be a multiple of 7.
  • minimum memory request of 256GB.
  • #PBS -q hugemem

GPU:

  • 2×12 cores (Intel Xeon Broadwell/Haswell) in 30 compute nodes
  • 4 x Nvidia Tesla 24 GBytes K80 Accelerator (or 8 x GPUs)) on each node
  • 256 GBytes of RAM on CPU
  • 700 GBytes of SSD local disk
  • charge rate, 3SU per cpuhour
  • #PBS -q gpu
  • #PBS -l ngpus = 2, minimum ngpus request is 2, in the multiple of 2, this is defined as the number of GPUs
  • #PBS -l ncpus = 6, minimum ncpus request is 6, in the multiple of 6, and 3 x ngpus

More details of the GPU specifications, how to use GPUs at NCI, and GPU enabled software are available here.

In Collaboration With